Estimated glomerular filtration rate slope and risk of primary and secondary major adverse cardiovascular events and heart failure hospitalization in people with type 2 diabetes: An analysis of the EXSCEL trial.

AIM: The decline in estimated glomerular filtration rate (eGFR), a significant predictor of cardiovascular disease (CVD), occurs heterogeneously in people with diabetes because of various risk factors. We investigated the role of eGFR decline in predicting CVD events in people with type 2 diabetes in both primary and secondary CVD prevention settings. MATERIALS AND METHODS: Bayesian joint modelling of repeated measures of eGFR and time to CVD event was applied to the Exenatide Study of Cardiovascular Event Lowering (EXSCEL) trial to examine the association between the eGFR slope and the incidence of major adverse CV event/hospitalization for heart failure (MACE/hHF) (non-fatal myocardial infarction, non-fatal stroke, CV death, or hospitalization for heart failure). The analysis was adjusted for age, sex, smoking, systolic blood pressure, baseline eGFR, antihypertensive and lipid-lowering medication, diabetes duration, atrial fibrillation, high-density cholesterol, total cholesterol, HbA1c and treatment allocation (once-weekly exenatide or placebo). RESULTS: Data from 11 101 trial participants with (n = 7942) and without (n = 3159) previous history of CVD were analysed. The mean ± SD eGFR slope per year in participants without and with previous CVD was -0.68 ± 1.67 and -1.03 ± 2.13 mL/min/1.73 m2, respectively. The 5-year MACE/hHF incidences were 7.5% (95% CI 6.2, 8.8) and 20% (95% CI 19, 22), respectively. The 1-SD decrease in the eGFR slope was associated with increased MACE/hHF risks of 48% (HR 1.48, 95% CI 1.12, 1.98, p = 0.007) and 33% (HR 1.33, 95% CI 1.18,1.51, p < 0.001) in participants without and with previous CVD, respectively. CONCLUSIONS: eGFR trajectories over time significantly predict incident MACE/hHF events in people with type 2 diabetes with and without existing CVD, with a higher hazard ratio for MACE/hHF in the latter group.

A New Family of Bivariate Exponential Distributions with Negative Dependence Based on Counter-Monotonic Shock Method.

We introduce a new family of bivariate exponential distributions based on the counter-monotonic shock model. This family of distribution is easy to simulate and includes the Fréchet lower bound, which allows to span all degrees of negative dependence. The construction and distributional properties of the proposed bivariate distribution are presented along with an estimation of the parameters involved in our model based on the method of moments. A simulation study is carried out to evaluate the performance of the suggested estimators. An extension to the general model describing both negative and positive dependence is sketched in the last section of the paper.